Highly active anti-retroviral therapy (HAART) can completely suppress viremia but does not lead to functional cure (HIV remission; i.e., no rebound viremia after stopping HAART). The Mississippi baby seemed exceptional. She was given HAART from 30 h after birth for 11/2 years and remained aviremic for 2.3 years off HAART. However, HAART was restarted recently due to recurrent viremia and seroconversion. In rhesus macaques (RMs), onset of HAART on day 3 post SIV inoculation prevented viremia and seroconversion (but not when started on later days). Although viremia was suppressed in all drug-treated RMs, it recurred in all animals after HAART was stopped at week 24. Together, these data indicate that HIV or SIV reservoirs are difficult to eradicate. We postulate that early-onset HAART - coupled with induction of strong T-cell immunity via multi-targeted live viral vaccine vectors will achieve functional cures and eradicate infectious virus from reservoirs. Functional cures will be manifested by lack of viral rebound when HAART is stopped. To probe for eradication, CD8+ cells will be depleted in RMs with viral remission; no infectious virus will reemerge. HAART + vaccination may also protect against subsequent virus challenges. We will test our hypothesis by combining HAART with a live attenuated vaccine vector that was safe and highly immunogenic in RMs: rubella virus expressing SIV Gag (rub-gag). When tested in RMs, rub-gag replicated vigorously, and the Gag insert was highly immunogenic. We plan to use the clade C simian-human immunodeficiency virus (SHIV-C) system, which has the advantage of allowing direct testing of immune responses against some HIV viral targets in RMs (e.g., Tat and Env). We will test our central hypothesis in the following Specific Aims: Aim 1: to test whether rub-gag immunization under HAART coverage will yield functional cures in infant RMs and to probe for SHIV-C eradication with CD8 depletion. HAART will be given from 48 h to week 24. Aim 2: to construct rubella/HIV Tat (rub-tat) vectors. We postulate that together with HAART, rub-gag + rub- tat will eradicate SHIV-C, as shown by CD8 depletion of RMs that show no rebound after stopping HAART. Aim 3: to test the longevity of the T-cell responses to rub-gag + rub-tat vaccine given under HAART coverage and to assess whether they protect the RMs against subsequent SHIV-C rechallenge. Aim 4: to test whether boosting with autologous virus will yield optimal immunity. These studies will parallel planned clinical trials in perinatally exposed infants who will be vaccinated with thir own inactivated virus while on HAART. The primate model will dissect mechanisms and probe viral reservoirs by CD8 depletion. Initial immunogenicity studies will be conducted in 4-month old RM infants that no longer need bottle feeding and permit collection of larger volumes of blood to analyze vaccine safety and immunogenicity. Once we have found a promising HAART/vaccine regimen that can achieve viral eradication in some of the treated RM infants, we will perform a more definitive study in neonatal RMs challenged orally with SHIV-C.